ARTICLE pubs.acs.org/JPCA
Rotational Spectrum of 1,1,1-Trifluoro-2-butanone Using Chirped-Pulse Fourier Transform Microwave Spectroscopy Luca Evangelisti,† Galen Sedo, and Jennifer van Wijngaarden* Department of Chemistry, University of Manitoba, Winnipeg, Manitoba R3T 2N2, Canada
bS Supporting Information ABSTRACT: The pure rotational spectra of 1,1,1-trifluoro-2butanone and its four 13C isotopologues have been studied using the new chirped-pulsed Fourier transform microwave spectrometer at the University of Manitoba in combination with a conventional Balle-Flygare-type instrument. Quantum chemical calculations, at the MP2/6-311þþG(d,p) level, were carried out to obtain information about the structure, relative stability, and difference in populations of the three lowest energy conformers corresponding to dihedral angles of 0, 82.8, and 119.2 along the carbon backbone. The observed spectra are that of conformer I (dihedral angle 0), and, based on analysis of the observed splitting, the V3 barrier to internal rotation of the methyl group has been determined to be 9.380(5) kJ mol-1. The spectroscopic constants of the five isotopologues were used to precisely derive the rs and partial r0 geometries of this conformer based on an assumed planar carbon backbone (as supported by the spectra and ab initio calculations).
’ INTRODUCTION In recent years, there has been increasing interest in exploiting the unusual properties of fluorocarbons to modulate physicochemical properties of molecules.1 Fluorine atoms are often introduced in a drug skeleton to modify pharmacokinetics properties. The rationale for such a strategy is that fluorine is a stronger electron-withdrawing group than other halogen atoms, and its size is similar to that of hydrogen. Despite their similar sizes, hydrogen and fluorine have quite different electronic properties as fluorine has a greater electronegativity and lower polarizability. Consequently, the carbon-fluorine bond is one of the strongest known in organic chemistry. To accurately tune the physiochemical properties of fluorinated organic molecules, it is critical to experimentally determine key properties related to molecular geometry and electronic structure.2-4 Microwave spectroscopy is a valuable technique5 for the study of conformational equilibria and internal motions of molecules, and it is well-known that even relatively small molecules can have very complicated rotational spectra. For example, the spectrum of acetylacetone (C5H8O2) is complex due to proton tunneling (keto-enol tautomerization) and a low barrier for internal rotation of the two methyl groups.6 Generally, the substitution of methyl groups by heavier rotating tops (for example, CF3) has a drastic effect on the pattern of the spectrum, and the electronic properties of these groups also influence which conformers or r 2011 American Chemical Society
tautomers are present. For example, hexafluoroacetylacetone (C5H2O2F6) is a rigid molecule in terms of internal rotation and tautomerization on the time scale of microwave spectroscopy.7 This rigidity arises due to a combination of factors. First, the reduced barrier to internal rotation is higher for the heavier CF3 moiety. This decreases the internal rotation splitting to values smaller than the resolving power of the spectrometer. Second, the proton tunneling splitting (as seen in the tautomerization of malonaldehyde,8 for example) is reduced by a factor larger than 5 106 in hexafluoroacetylacetone due to inertial effects as it requires the simultaneous rotation of both CF3 groups.7 Moreover, the high electronegativity of the fluorine atoms seems to increase the distance between the two oxygen atoms according to theoretical investigations.9 The rotational spectrum of trifluoroacetone10 was recently explored, and its internal rotor dynamics was compared to that of acetone.11 In the fluorinated species, only internal rotation of the CH3 moiety was observed, and its barrier (3.28 or 3.10 kJ mol-1 by the principal axis method (PAM)12 and combined axis method (CAM),13 respectively) was found to be surprisingly similar to that of acetone (3.27 kJ mol-1). Note that the experimentally derived values vary as a consequence of the Hamiltonians Received: September 20, 2010 Revised: November 8, 2010 Published: January 10, 2011 685
dx.doi.org/10.1021/jp1089905 | J. Phys. Chem. A 2011, 115, 685–690
The Journal of Physical Chemistry A
ARTICLE
used in the PAM and CAM methods as described in reference.10 On the basis of this result, it was concluded that the strong electron-withdrawing tendency of the CF3 group did not have a large effect on the V3 barrier of the CH3 subunit. In this Article, we extend this investigation of fluorinated molecules to the study of 1,1,1-trifluoro-2-butanone (TFB) for comparison with 2-butanone. The thermodynamic and spectroscopic properties of 2-butanone have been extensively studied.14-18 The microwave spectrum was used to identify the lowest energy conformer and to derive the barriers to internal rotation of the two methyl groups. To the best of our knowledge, there is no comparable microwave study of TFB. Our ab initio calculations suggest that there are multiple conformers of TFB that correspond to different angles along the carbon backbone (dihedral angles of 0 (conformer I), 82.8 (conformer II), and 119.2 (conformer III)). All three conformers of TFB have large dipole moment components, and thus TFB provides an ideal prototype molecule for testing our new broadband microwave spectrometer.
generator and covers (up to) a 6 GHz total bandwidth. The broadband pulse is amplified (6) using a 5 W solid state amplifier that employs a fast (ns) TTL controlled switch and is then coupled into the custom-built vacuum chamber (d = 18 in., l = 36) lined with microwave absorbing foam. The chamber houses a pair of high gain horn antenna (7) for broadcasting and receiving the microwave signal. The vacuum chamber is evacuated by a 12 in. diffusion pump (Varian VHS 10) and two rotary vane pumps (Varian 3 phase DS602) to achieve a background pressure of ∼8 10-7 Torr. The molecular sample is introduced into the cavity via a supersonic jet expansion of a gas mixture seeded with the molecule of interest using a pulsed solenoid valve (8) with an exit diameter of 1.0 mm. The nozzle is mounted perpendicularly to the propagation of the microwave radiation and faces directly into the throat of the diffusion pump. The detection branch of the microwave circuit is protected during the excitation pulse using a diode limiter (9) ahead of a SPST switch (10). After the polarizing pulse has dissipated, the weak molecular emission is allowed through the circuit to the low noise amplifier (11) and frequency downconverted using a second microwave signal generator (4b) and mixer (12). The emission signal is filtered (13, 14) and digitized using a 6 GHz bandwidth oscilloscope (15). As all three signal generators and the oscilloscope are referenced to a 10 MHz rubidium standard (16), the experimental sequence may be repeated, and individual FID events may be phase coherently averaged over several hours. A fast Fourier transformation (FFT) is applied to the resulting signal to obtain a spectrum over the full bandwidth of the chirp. The typical experimental repetition rate is 10 Hz and is limited by the vacuum throughput and/or digitization. The trigger for the experiment, the timing for the switches (6, 10), and the pulsed nozzle are controlled by a four-channel pulse delay generator. A commercial sample of TFB (purchased from Aldrich) was used without further purification. The carrier gas (Ne and He) was passed over the TFB at room temperature, and the spectra of the all 13C species were measured in natural abundance. Typical linewidths in the cp-FTMW spectrum were ∼200 kHz, although this can be reduced by lengthening the FID acquisition beyond the 10 μs used in these experiments. The transition frequencies were verified (and more transitions were sought) using a Balle-Flygare-type Fourier transform microwave (FTMW) spectrometer, which has been described elsewhere.21 As the supersonic jet is arranged coaxially to the cavity axis in this second instrument, each rotational transition is split by Doppler effect. The estimated accuracy of frequency measurements is (1 kHz, and the typical linewidths (fwhm) are 7 kHz with this instrument.
’ EXPERIMENTAL SECTION The rotational spectrum of 1,1,1-trifluoro-2-butanone was first recorded using the recently built chirped-pulse FTMW (cp-FTMW) spectrometer at the University of Manitoba. This new instrument operates from 8 to 18 GHz and follows the earlier design of Pate and co-workers19 with later modifications by Cooke and co-workers.20 The cp-FTMW technique provides broadband access to the microwave spectral region with more reliable intensity information across the range of the spectrometer. The instrument design of our new spectrometer is described briefly here, and the numbers in boldface follow the schematic in Figure 1. A chirped pulse is created using a 6 GSa/s arbitrary waveform generator (1) that employs a 6 GHz phase-locked dielectric resonance oscillator (PLDRO) (2) as a stable external clock. Typical waveforms used are linear frequency sweeps through 1-3 GHz in the span of 1-5 μs. The pulse is filtered (3) and combined with the cw output of a microwave signal generator (4a) using a double balanced mixer (5). The resulting pulse is centered at the output frequency of the microwave signal
Figure 1. Schematic of the 8-18 GHz bandwidth chirped-pulse Fourier transform microwave spectrometer: (1) 6 GSa/s arbitrary waveform generator. (2) Phase-locked dielectric resonance oscillator. (3) Low pass filter. (4a) Microwave signal generator. (4b) Microwave signal generator. (5) Double balanced mixer. (6) 5 W solid state amplifier. (7) High gain horn antennae. (8) Pulsed nozzle valve. (9) Diode limiter. (10) SPST switch. (11) Low noise amplifier. (12) Mixer. (13) Low pass filter. (14) DC block. (15) Digital oscilloscope. (16) 10 MHz rubidium standard.
’ THEORETICAL CALCULATIONS Before the initial search for rotational transitions, quantum chemical calculations were completed using the Gaussian 03 software package22 to identify possible conformers. For each potential conformer, we constructed a structural Z-matrix and performed a full geometry optimization at the Hartree-Fock self-consistent field level of theory using the 6-311G(d,p) basis set. We then refined these calculations by including electron correlation effects using Møller-Plesset terms up to second order [MP2/6-311þþG(d,p)]. In total, three energy minima were found and verified by subsequent harmonic frequency calculations. These minima correspond to the structures shown at the top of Table 1 (with dihedral angles of the carbon backbone 0 686
dx.doi.org/10.1021/jp1089905 |J. Phys. Chem. A 2011, 115, 685–690
The Journal of Physical Chemistry A
ARTICLE
Table 1. MP2/6-311þþG(d,p) Spectroscopic Parameters and Relative Energies of TFB
a This species is calculated to be slightly nonplanar, with a Cs barrier of about 0.056 kJ mol-1. b Absolute energy -529.099737Eh. c Absolute energy -529.008883Eh.
(conformer I), 82.8 (conformer II), and 119.2 (conformer III)) with conformer I predicted to fall lowest in energy. The relative energies derived from the MP2 calculations provide an indication of the relative populations of the conformations before supersonic expansion. In addition, the ab initio calculations provided an estimate of the rotational constants, dipole moment components along the principal inertial axes, and the angles — (a,i) that the internal rotation axes of the methyl and trifluoromethyl groups form with the a-principal axis of the three conformers. For the lowest energy structure, the barriers for the internal rotation of the methyl and trifluoromethyl groups were also predicted. Starting from the geometry for conformer I, the orientation of the CH3 (and later the CF3) moiety was changed (by rotating about its C3 axis), and the positions of all atoms were reoptimized at each step with the exception of the HCCC (and later the FCCC) dihedral angle.
’ SPECTRAL ASSIGNMENT AND ANALYSIS On the basis of the preliminary calculations of the rotational constants, the spectrum of the lowest energy conformer was predicted. Very intense μa-R-type transitions, with Ka = 0-3, and with J ranging from 4 to 6, were readily identified and assigned using the cp-FTMW spectrometer followed by assignment of the less intense μb-R-type transitions. A sample spectrum covering 2 GHz is shown in Figure 2. After assignment of the main conformer, a number of weak lines remained unassigned in the broadband spectrum. These transitions were soon attributed to the less abundant 13C (∼1%) istopologues. In total, the spectra of all four 13 C isotopologues were assigned, and the frequencies of all species are listed in the Supporting Information. All transition frequencies were verified using the Balle-Flygare FTMW instrument. With the higher resolution of the Balle-Flygare spectrometer, a careful analysis of the spectra revealed that a number of transitions have a complex structure arising from the internal rotation of the methyl group furthest from the carbonyl group. The sublevels generated by the internal rotation are classified according to their symmetry as A and E.5 Correspondingly, the rotational transitions are split into doublets, using the same A/E labeling scheme as no transition is allowed between states of different symmetry. The transition 63,4-53,3 is shown in Figure 3 as an example of the splitting observed for the higher Ka transitions. Many of these weaker transitions appeared broadened (15 kHz fwhm in this case), but the A/E splitting was readily resolved with the FTMW instrument.
Figure 2. 2 GHz portion of the broadband rotational spectrum of TFB as collected by the cp-FTMW after 50 000 signal averages (top trace) and the theoretical spectrum (bottom trace) based on the calculated rotational constants.
For the normal species, a global fit of all component lines was performed with the computer program XIAM,13 which uses the combined axis method (CAM), using the S reduction of Watson’s quartic Hamiltonian in the Ir representation.23 The fitting results are a “rigid” limit set of rotational and first-order centrifugal distortion constants, common to A and E sublevels, together with the V3 barrier, while the IR moment of inertia of the methyl group and the angle — (a,i) were fixed to the values determined from the theoretical geometry. The spectroscopic constants for the normal species obtained with this approach are reported in Table 2. As the transitions of the 13C isotopologues were extremely weak, the internal rotor splitting was not measured. The observed lines were fit using Watson’s semirigid Hamiltonian in the S reduction and Ir representation23 in SPFIT Pickett’s program.24 The resulting spectroscopic constants derived for all four 13C isotopologues are reported in Table 3.
’ DISCUSSION A. Molecular Structure. By comparison of the experimental rotational constants of Table 2 to the predicted ab initio 687
dx.doi.org/10.1021/jp1089905 |J. Phys. Chem. A 2011, 115, 685–690
The Journal of Physical Chemistry A
ARTICLE
Table 3. Experimental Spectroscopic Constants Derived for the Observed 13C Isotopologues of TFB (S Reduction, Ir Representation) SPFIT
13
13
C1
C2
a
13
C3
13
C4
A/MHz
3517.371(2)
3514.013(2)
3503.625(2)
3515.757(2)
B/MHz
1398.5995(3)
1400.5445(3)
1392.1053(3)
1367.3531(3)
C/MHz
1235.4214(3)
1236.5564(3)
1228.7192(3)
1210.8160(3)
DJ/kHz
0.167(4)
0.168(4)
0.167(4)
0.157(4)
DJK/kHz d2/kHz
0.37(6) 0.055(7)
0.38(6) 0.053(7)
0.48(6) 0.056(7)
0.44(6) 0.054(7)
σ/kHzb Nc
1
2
2
2
23
23
23
23
a
Error in parentheses in units of the last digit. b Root-mean-square deviation of the fit. c Number of lines in the fit. Figure 3. A single rotational transition of TFB (63,4-53,3) illustrating the A and E components due to the internal rotation of the methyl group (recorded using the Balle-Flygare FTMW spectrometer).
Table 2. Experimental Spectroscopic Constants Derived for the Observed Normal Isotopologue of TFB (S Reduction, Ir Representation) XIAM
normal
A/MHz
3517.2267(2)a
B/MHz
1401.7179(2)
C/MHz
1237.8592(1)
DJ/kHz
0.171(2)
DJK/kHz
0.377(7) -0.013(1)
d1/kHz d2/kHz V3/kJ mol-1
0.047(1) 9.380(5)
IR/uÅ2
3.18b
— (a,i)/deg
c
σ/kHz
d
Ne
Figure 4. The atom numbering convention of TFB in this work.
19.89b 3 97
a Error in parentheses in units of the last digit. b Fixed to the value of the ab initio calculation. c The value of — (b,i) is the complement of — (a,i), and the value of — (c,i) is 90 (from the planarity of the mainframe). d Root-mean-square deviation of the fit. e Number of lines in the fit.
Figure 5. Ab initio (MP2/6-311þþG(d,p)) potential curve for the interconversion between the three conformers of TFB. The three minima from left to right correspond to conformers I, II, and III, respectively.
theoretical values of Table 1, we can conclude that the observed spectrum corresponds to conformer I. Conformers II and III were not detected despite the fact that their relative populations are 40% and 20% that of conformer I (at room temperature), respectively. The failure to observe the higher energy forms is probably due to the conformational relaxation process around the dihedral angle O5C2-C3C4 as shown in Figure 4. To estimate the potential curve for interconversion between the conformers, an ab initio scan was carried out in steps of 2 over the full range of the dihedral angle of the carbon (and in steps of 1 close to the maxima). While the dihedral angle was kept fixed at every step, the rest of geometric parameters were reoptimized for each point along the path. The results are plotted in Figure 5, and the low interconversion barriers (0.31 and 0.78 kJ mol-1) support the assertion that the two higher energy conformers relax into the lowest energy conformer during supersonic expansion.25
The inertial defect5 of TFB is calculated to be -95.96 uÅ2. This is similar to the value derived for trifluoroacetone (-92.64 uÅ2),10 with the main difference arising due to the contributions of the two out-of-plane hydrogen atoms from the CH2 group. A similar difference, for example, is found upon comparison of the inertial defects of acetone (-6.276 uÅ2)11 and 2-butanone (-9.45 uÅ2).16 This suggests that TFB has an approximately planar mainframe as reported for 2-butanone. The ab initio calculations performed in this work suggest a small nonplanarity of the mainframe, but the barrier to planarity falls below the vibrational ground state. If the backbone geometry is constrained to be planar in the calculations, the value of the inertial defect is -95.992 uÅ.2 On the basis of these comparisons, we assert that mainframe of 688
dx.doi.org/10.1021/jp1089905 |J. Phys. Chem. A 2011, 115, 685–690
The Journal of Physical Chemistry A
ARTICLE
Table 4. Structural Parameters of TFB: The Experimental Substitution Coordinatesa
Table 5. Effective (r0) Structure of TFB, Based on the MP2/ 6-311þþG(d,p) Geometrya bond distances (Å)
substitution coordinates/Å |as|
|bs|
13
-0.898 (2)
0.08i (2)
13
0.550 (3)
0.363 (4)
13
1.579 (1) 3.017 (1)
-0.753 (2) -0.250 (6)
C1 C2
C3 13 C4
b
Calculated by fixing |c| to zero for these atoms. Note that while the Kraitchman equations provide only absolute values of the coordinates, the signs are inferred on the basis of the orientation of the conformer in its principal axis system based on the ab initio predictions. b Imaginary value. The rs values of the bond distance and the valence angles were: C2-C1 = 1.493(6); C3-C2 = 1.517(4); C4-C3 = 1.524(3); C3C2C1 = 118.6(6); C4C3C2 = 113.4(4). Calculated by fixing to zero the |bs| value of C1. a
conformer I is effectively planar, and the molecule has a symmetry plane containing the ab axes. This is further supported by the fact that we did not observe c-type transitions for this conformer. With five isotopologues characterized, the resulting 15 groundstate rotational constants were used to derive structural information about this conformer as described below. i. rs Position of the Four Carbon Atoms. By applying Kraitchmann’s equations,26 we obtained the coordinates of the heavy atoms and subsequently the bond lengths and valence angles, as listed in Table 4. The b-coordinate of C1 resulted in a small imaginary number. This is likely a result of vibrational effects involving the trifluoromethyl group. The uncertainties reported for the substitution structure coordinates were calculated according to the Constain rule.27 ii. r0 Geometry. We obtained a partial r0 geometry of the molecule, limited to the determination of the structural parameters connecting the heavy atoms (C and F), from the fit of the 15 rotational constants of the vibrational ground states of the five isotopologues. We allowed one bond length (C2-C1), one valence angle of the molecular frame (C3C2C1), and the (F8)F7C1-C2O5 dihedral angle to change, with respect to the ab initio values, by allowing “confidence intervals” of 0.01 Å for the bond distance and of 1 for the valence angles, respectively, according to the diagnostic least-squares procedure described by Curl.28 The geometric parameters chosen for the leastsquares fitting routine are those that have the most significant effect on the rotational constants (i.e., largest derivatives dBo/ d(angle) or dBo/d(bond length), for example). The results are shown in Table 5 and are compared to the re structure of the theoretical equilibrium values. The maximum discrepancy between the observed and calculated rotational constants is 0.2%. The structural parameters of the remaining atoms were fixed to the MP2/6-311þþG(d,p) values as reported in Table 5. B. Internal Rotation of CH3. As previously reported for 2-butanone,16-18 the TFB spectra exhibit a doublet fine structure with a distinct splitting between components of A- and E-type symmetry. This is characteristic of molecules with a single methyl rotor and a relatively low potential barrier to internal rotation.5 Upon comparison of the internal rotation of the methyl group furthest from the carbonyl subunit, the V3 barriers of 2-butanone (9.51 ( 0.12 kJ mol-1)18 and TFB (9.38 ( 0.05 kJ mol-1) agree within the experimental uncertainties, and the value for TFB is considerably smaller (15%) than the theoretical value of
valence angles (deg)
dihedral angles (deg)
b
C2-C1
1.542(1)
C3-C2 C4-C3
1.509 1.524
C3C2C1 C4C3C2
116.1(2)b 112.3 C4C3-C2C1
O5-C2
1.212
O5C2C3
125.4
O5C2-C3C4
0.0
F6-C1
1.326
F6C1C2
112.5
F6C1-C2O5
0.0
F7-C1
1.347
F7C1C2
109.9
F7C1-C2O5
120.5(1)b
F8-C1
1.347
F8C1C2
109.9
F8C1-C2O5
-120.5(1)b
H9-C4
1.092
H9C4C3
110.8
H9C4-C3C2
180.0
-59.8
H10-C4 1.092
H10C4C3 110.8
H10C4-C3C2
59.8
H11-C4 1.092 H12-C3 1.097
H11C4C3 110.1 H12C3C2 107.5
H11C4-C3C2 H12C3-C2C1
180.0 56.7
H13-C3 1.097
H13C3C2 107.5
H13C3-C2C1
-56.7
The determined parameters in bold were fit (uncertainty in parentheses) to reproduce the rotational constants. b The MP2/6-311þþG(d,p) values of these parameters were C2C1 = 1.548 Å, C3C2C1 = 115.2, F7(F8)C1-C2O5 = (120.9. a
11.1 kJ mol-1 [MP2/6-311þþG(d,p)]. The comparison with the ab initio predictions highlights the importance of accurate experimental measurements of such parameters. As reported earlier for trifluoroacetone,10 no splittings due to the CF3 internal rotation were observed for TFB. This is attributed to the larger moment of inertia of this internal rotor and its relatively high reduced barrier.
’ CONCLUSIONS In this work, we observed and assigned the rotational spectra of the normal species of 1,1,1-trifluoro-2-butanone and its four 13 C isotopic species for the first time using our new cp-FTMW spectrometer. The rotational constants were used to derive the structure of the organic mainframe. Our results determined that the lowest energy form of TFB was conformer I analogous to that reported for 2-butanone. The barrier to CH3 internal rotation was similar in the two species, suggesting that the exchange of CH3 with the electron-withdrawing CF3 moiety on the other end of the molecule has little effect on the internal motion. The results demonstrate, once again, that pulsed-jet FTMW spectroscopy is a valuable tool for extracting precise information on the structure and hindered internal motions of molecules. ’ ASSOCIATED CONTENT
bS
Supporting Information. Table of transition frequencies. This material is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION Corresponding Author
*Tel.: (204) 474-8379. Fax: (204) 474-7608. E-mail: vanwijng@ cc.umanitoba.ca. Present Addresses †
Dipartimento di Chimica “G. Ciamician” dell’Universita, Via Selmi 2, I-40126 Bologna, Italy.
689
dx.doi.org/10.1021/jp1089905 |J. Phys. Chem. A 2011, 115, 685–690
The Journal of Physical Chemistry A
ARTICLE
’ ACKNOWLEDGMENT We would like to thank the University of Bologna for financial support to allow L.E. to conduct experiments at the University of Manitoba. We are extremely grateful for the generous guidance of our microwave colleagues S. Shipman (New College of Florida), B. H. Pate (U. Virginia), and S. A. Cooke (U. North Texas) during the design and construction of our new cp-FTMW spectrometer. Funding for the new instrument was provided by the Canada Foundation for Innovation (CFI) Leaders Opportunity Fund, and ongoing research support for this work is provided by the Natural Sciences and Engineering Research Council (NSERC) of Canada through the Discovery Grant and University Faculty Award programs. L.E. wishes to thank Professor W. Caminati for his helpful teachings. ’ REFERENCES (1) Lee, H.-Y.; Lee, K. H.; Al-Hashimi, H. M.; Marsh, E. N. G. J. Am. Chem. Soc. 2006, 128, 337. (2) Howard, J. A. K.; Hoy, V. J.; O’Hagan, D.; Smith, G. T. Tetrahedron 1996, 52, 12613. (3) Jacobsen, H. Phys. Chem. Chem. Phys. 2009, 11, 7231. (4) Schaal, H.; H€aber, T.; Suhm, M. A. J. Phys. Chem. A 2000, 104, 265. (5) Gordy, W.; Cook, R. L. Microwave Molecular Spectra, 3rd ed.; Wiley: New York, 1984. (6) Caminati, W.; Grabow, J.-U. J. Am. Chem. Soc. 2006, 128, 854. (7) Evangelisti, L.; Tang, S.; Velino, B.; Giuliano, B. M.; Melandri, S.; Caminati, W. Chem. Phys. Lett. 2009, 473, 247. (8) Baughcum, S. L.; Smith, Z.; Wilson, E. B.; Duerst, R. W. J. Am. Chem. Soc. 1984, 106, 2260. (9) Hargis, J. C.; Evangelista, F. A.; Ingles, J. B.; Schaefer, H. F., III. J. Am. Chem. Soc. 2008, 130, 17471. (10) Evangelisti, L.; Favero, L. B.; Maris, A.; Melandri, S.; Vega-Toribio, A.; Lesarri, A.; Caminati, W. J. Mol. Spectrosc. 2010, 259, 65. (11) Swales, J. D.; Constain, C. C. J. Chem. Phys. 1959, 31, 1562. (12) Wilson, E. B. Chem. Rev. 1940, 27, 17. (13) Hartwig, H.; Dreizler, H. Z. Naturforsch. 1996, 51a, 923. (14) Chao, J.; Zwolinski, B. J. J. Phys. Chem. 1976, 5, 319. (15) Durig, J. R.; Feng, F. S.; Wang, A.; Phan, H. V. Can. J. Chem. 1991, 69, 1827. (16) Pierce, L.; Chang, C. K.; Hayashi, M.; Nelson, R. J. Mol. Spectrosc. 1969, 32, 449. (17) Mamleev, A. Kh.; Gunderova, L. N.; Dronov, V. I.; Pozdeev, N. M. Khim. Fiz. 1982, 309. (18) Pozdeev, N. M.; Mamleev, A. Kh.; Gunderova, L. N.; Galeev, R. V. J. Struct. Chem. 1988, 29, 52. (19) Brown, G. G.; Dian, B. C.; Douglass, K. O.; Geyer, S. M.; Shipman, S. T.; Pate, B. H. Rev. Sci. Instrum. 2008, 79, 053103. (20) Grubbs, G. S.; Dewberry, C. T.; Etchison, K. C.; Kerr, K. E.; Cooke, S. A. Rev. Sci. Instrum. 2007, 78, 096106. (21) Sedo, G.; van Wijngaarden, J. J. Chem. Phys. 2009, 131, 044303. (22) Frisch, M. J.; et al. Gaussian 03, revision B.04; Gaussian, Inc.: Pittsburgh, PA, 2003. (23) Watson, J. K. G. In Vibrational Spectra and Structure; Durig, J. R., Ed.; Elsevier: New York/Amsterdam, 1977; Vol. 6, pp 1-89. (24) Pickett, H. M. J. Mol. Spectrosc. 1991, 148, 371; http://spec.jpl. nasa.gov. (25) Fraser, G. T.; Suenram, R. D.; Lugez, C. L. J. Phys. Chem. A 2000, 104, 1141. (26) Kraitchman, J. Am. J. Phys. 1953, 21, 17; http://www.ifpan.edu. pl/∼kisiel/struct.htm#kra. Implemented using KRA suite program available from Z. Kisiel. (27) Constain, C. C. Trans. Am. Crystallogr. Assoc. 1966, 2, 157. (28) Curl, R. F. J. Comput. Phys. 1970, 6, 367. 690
dx.doi.org/10.1021/jp1089905 |J. Phys. Chem. A 2011, 115, 685–690
